The Indus Civilization Read online

  A South Asian center for the domestication of plants and animals is an old concept, going back to the famous Russian botanist Nicoli Vavilov.3 A number of scholars have observed that the Afghan-Baluch region is environmentally and ecologically very much akin to the entire Iranian Plateau and the uplands of the regions bordering the Mediterranean: It has a steppe-like quality with pistachio, juniper, and almond tree cover, along with the hard, cold winters in which wheat and barley evolved.4 It is also within the range of the winter westerlies, which bring moisture, often in the form of snow, to the Near East on across the Iranian Plateau to the Punjab and western Sindh.5 What this tells us is that the Afghan-Baluch region is a perfectly reasonable place for both wild barley (which is documented) and wild wheat to have been found. The absence of wild wheat there today may mean that the modern distribution of the wild ancestors of early domesticates may not reflect the late glacial or early Holocene distribution of the same species. In other words, the absence of wild wheat in the Afghan-Baluch region in modern times does not mean that it was not there in deep antiquity. Given the approximately 10,000 years of intensive land use in this region, wild wheat could have been on the regional “endangered species list” many thousands of years ago and may have ended up extinct in the region. This observation clearly needs serious research before it is accepted, but it is an attractive hypothesis because of (1) the general nature of the Afghan-Baluch environment, and (2) the presence of wild barely as well as other early domesticates, especially the animals.

  The presence or absence of wild wheat in the Afghan-Baluch region is important because without wheat the case for an indigenous process of domestication in the region is difficult to make. As the story of early domestication in western South Asia unfolds, particular attention is paid to this grain.

  Early Food Production and Domestication

  It has been virtual archaeological dogma for decades that Braidwood’s constellation of potentially domesticable plants and animals (wheat, barley, sheep, goats, cattle, pigs) were first domesticated in the Near East (Israel, Lebanon, Syria, southwestern Turkey, Iraq, western Iran) early in the Holocene (c. 8,000 to 10,000 years ago). The most coherent story has been told for the Greater Jordan Valley and the sequence from the sophisticated Natufian hunter-gatherers through “Pre-Pottery Neolithic A” (PPNA) and “Pre-Pottery Neolithic B” (PPNB). The chronology is given in table 2.1.

  Table 2.1 A chronology for food production in the Greater Jordan Valley

  PPNB 7300/7200 to c. 5800 B.C.

  PPNA Sultanian 8300/100 to 7300/200 B.C.

  PPNA Khiamain 8500/8300 to 8100 B.C.

  Natufian Late c. 9000 to 8500/8300 B.C.

  Natufian Early 10,800/10,500 B.C. to c. 9000 B.C.

  Geometric Kebaran 12,000 to 10,800/ 10,500 B.C.

  After Bar-Yosef and Belfer-Cohen 1989:455.

  The sequence can be briefly summarized. The Natufians were advanced hunter-gatherers, some of whom were sedentary for most, perhaps all, of the year. Domesticated plants are associated with PPNA villages, such as Jericho. Domesticated animals, at least the biological features that allow archaeozoologists to identify them, come later, in PPNB villages.6

  The usual story is that domesticated plants and animals, and the techniques of food production, then somehow “diffused” to other parts of the Old World, including South Asia. Some scholars are quite explicit about this matter.7 Those who hold to this view can point to the early dates from Near Eastern sites and the robust archaeological data sets that backstop their position. There is no doubt that there are early dates for food-producing sites in the Near East and that the substantial amount of excavation there has yielded a coherent culture historical sequence, but it has led to a kind of self-fulfilling prophesy. Convinced that the Near East was the early center, archaeologists have turned their attention to the investigation of this region, at the expense of others.

  It is now possible to mount a challenge to this archaeological dogma about the domestication of plants and animals. The data come from Afghanistan and Mehrgarh on the Kachi Plain of Pakistan.

  THE ORIGINS AND DEVELOPMENT OF FOOD-PRODUCING PEOPLES IN SOUTH ASIA

  The western borderlands of southern Asia are an excellent environment in which to conduct archaeological research on food production and domestication. The raw materials are there in the form of wild ancestors of early domesticates. Moreover, this region has an abundance of archaeological remains that have barely begun to be exploited. That there are thousands of unrecorded sites is a probability.

  Experiments with Food Production and Domestication in Afghanistan

  In the late 1960s and early 1970s A. Vinogradov conducted a survey bordering the Amu Darya.8 This survey produced masses of microlithic stone tools, implying the significant presence of early Holocene hunter-gatherers within a zone that has potentially domesticable resources. This provides a tie to two sites on the Balkh River in the northern hills of Afghanistan at the town of Aq Kupruk (figure 2.1).

  L. Dupree conducted small-scale excavations at the sites of Horse Cave (Aq Kupruk II) and Snake Cave (Aq Kupruk I) on the Balkh River in Afghanistan.9 Horse Cave has evidence for domesticated sheep and goats at circa 10,000 B.C. This follows an Upper Paleolithic occupation. This date is somewhat corroborated by the evidence from Snake Cave across the Balkh River, where there is evidence for domesticated sheep and goats at circa 7500 B.C.10

  The materials from these two extraordinary sites were never fully analyzed, so these observations must be considered preliminary in nature. But they may be a telltale sign of something important and should not be dismissed simply because they challenge the dogma of the Near Eastern center for the domestication of barley, wheat, cattle, sheep, and goats. The Aq Kupruk caves may give us reason to believe that important innovations leading to the domestication of plants and animals in human subsistence practices were taking place in areas other than the Near East. More and better data on the early domestication of plants and animals come from Mehrgarh in Pakistan.

  Figure 2.1 Some early sites in South Asia with evidence for food production

  Mehrgarh: An Early Village Farming Community

  The early food-producing village of Mehrgarh is located on the Kachi Plain of the Indus Valley. Kachi is an extension of the Indus plain into an anomalous “nick” in the eastern edge of the Iranian Plateau. It is thought that the Pleistocene Indus River flowed in this area, well to the west of its present course, so the alluvium is quite deep.11 The Bolan River provides a major route of communication between the Indus Valley and Baluchistan. It is the principal hydrological feature of the Kachi Plain today where it runs along the eastern edge of Mehrgarh. The Bolan Pass is a relatively easy route to the Quetta Valley and central Baluchistan. Mehrgarh sits at a strategic place, at the base of this route, just off the central plain of the Indus River. This is a very important, special location; a hub of communication, a place where peoples met and mixed.

  Mehrgarh Period I

  There is an excellent book on Mehrgarh that is well illustrated and tells the story of the first eleven seasons of excavation there.12 Period I was found in the northeastern corner of the site in excavation area MR 3 (figure 2.2). It can be seen from the plan that Mehrgarh is not a “tell” with levels stacked one on the other. Instead, the settlers here occupied one portion of the bank of the Bolan River for some centuries and then shifted their settlement pretty consistently to the south. So, Mehrgrah is a linear site, more than 2 kilometers long and over 1 kilometer wide in places. The stratigraphy of the site also is immensely complex, but it seems that Period I is aceramic, with pottery first appearing in Period II. Dwellings, frequently subdivided into four or even six rooms, were found made of simple mud brick (figure 2.3). The floors of these houses occasionally have the impression of reeds. Ovens and hearths were usually found in the corners of rooms, and signs of their use can be seen as traces of smoke on the plastered walls. One circular oven was lined with bricks and had a dome.

  Fig
ure 2.2 Plan of Mehrgarh (after Jarrige, Jarrige, Meadow, and Quivron 1995)

  Bricks of regular size were the standard. They are generally bun-shaped and have finger impressions on their tops. Some house walls were thin, with only one course of bricks; others were wider, with two or three. There were also compartmentalized storage structures (figure 2.4). These buildings may have been entered through the roof since no doors have been found and some walls are preserved to a substantial height; one to a distance of 3 meters.

  The technology of Mehrgarh I was relatively simple. There is native copper. The bed of the Bolan River carries cobbles of light brown flint, from which tools were fashioned: Some were made into sickle blades. A profusion of ground-stone food-processing tools was found, including a large number of quern and grinding-stone fragments. Baskets are in evidence, some of which were lined with bitumen to strengthen and waterproof them. There is evidence for the manufacture of calcite beads. Shell bangles and clay figurines were also present.

  Figure 2.3 House of Mehrgarh Period I (after Jarrige, Jarrige, Meadow, and Quivron 1995)

  A great deal of information on the paleobotany of Mehrgarh is available. The collection from Period I is especially rich. The dominant plant of Period I is domesticated, naked six-row barley. There are two other varieties of domesticated barley as well. Domesticated wheat is present in the form of einkorn, emmer, and a free-threshing hard durum, but in amounts much smaller than the barley sample. The noncereals so far identified for the period include the Indian jujube and dates, represented by stones in Periods I and II. Richard Meadow has observed: “While there is a possibility that local wild barleys could have been brought under cultivation in the Mehrgarh area, this is much less likely for wheat.”13 On the other hand there are some hints of a process that could be pursued. The free-threshing durum wheat can be seen as arising from a cross between emmer and goat face grass, which is found in Baluchistan. Thus, the conditions for the emmer—goat face grass cross existed locally, and free-threshing durum wheat ”could have originated as easily in the Pakistan area as in northwestern Iran.“14

  The animal economy is dominated by twelve species of what can be termed “wild big game”: gazelle, swamp deer, nilgai, blackbuck, onager, spotted deer, water buffalo, sheep, goats, cattle, pigs, and elephants. These are animals that would have lived on the Kachi Plain itself and the hills that surround it. “The lack of substantial numbers of fish and bird remains suggests that the Bolan Pass and the lake/swampy environments associated with it were of little importance to them,”15 but no screening was undertaken at Mehrgarh, and the recovery of fish and bird bone was therefore somewhat compromised.

  Meadow has noted that it is now clear that cattle (the humped South Asian zebu) were locally domesticated at Mehrgarh.16 He goes on to note:

  1. Sheep are likely to have been domesticated from local wild stock during Period I.

  2. Goats were kept from the time of the first occupation of the site.

  3. Size diminution in goats was largely complete by late Period I, in cattle by Period II, and in sheep perhaps not until Period III.

  4. The contribution of domestic or “pro-domestic” stock to the faunal assemblages came to surpass that of other animals early in the aceramic, but not in the earliest levels.

  6. The development of animal keeping by the ancient inhabitants of Mehrgarh took place in the context of cereal crop cultivation, the building of substantial mud-brick structures, and the existence of social differentiation and long-distance trade networks as attested by the presence of marine shells, lapis lazuli, and turquoise in even the earliest graves.17

  Figure 2.4 Compartmented buildings of Mehrgarh I and II (after Jarrige, Jarrige, Meadow, and Quivron 1995)

  The View Taken Here: Expanding the Geographical Dimensions of the Nuclear Zone

  Most (perhaps all) of the wild progenitors of the plants and animals on which premodern western and southern Asian civilizations are founded are native to the uplands of the Near East, Iran, and the Afghan-Baluch region. From the point of view of soil, water, and climate, these regions are suitable for growing wheat and barley and raising cattle, sheep, and goats on a significant scale. Thus, the environment is right for the early domestication of these plants and animals, and Meadow has carefully documented the fact that zebu cattle were a product of local domestication.

  One of the physical changes that took place as cattle, sheep, and goats were domesticated was size reduction. This was a progressive change that took centuries. Meadow documents progressive size reduction and, therefore, what seems to be the process of domestication for not only cattle, but sheep and goats as well. This is nicely seen in graphs he has prepared and implies that the inhabitants of Mehrgarh were manipulating these animals in the same way that their neighbors all across the Iranian Plateau to the Mediterranean Sea were.18 Since this is a gradual process involving centuries, the fact that Meadow can document it at Mehrgarh is good evidence that the domestication process was a local one.

  There are two pieces of evidence that inform us that domesticated cattle, sheep, and goats probably were not imported to Mehrgarh. First, the cattle are the local zebu breed, not the taurine cattle of the west. Second, the cattle, sheep, and goats at Mehrgarh start out as wild beasts and are gradually brought under domestication. The case is so strong for the local domestication of cattle that this part of the discussion can be set aside. There is a chance, one might suppose, that partially domesticated sheep and goats were continuously migrating to Mehrgarh from the west, somehow replacing or gradually interbreeding with the local stock not disrupting the size reduction documentation. This would have had to be a rather steady process, always unidirectional, west to east, from the culturally advanced Near East into “primitive” South Asia. Such a model is intellectually quite inefficient and filled with old-fashioned notions about the cultural “supremacy” on the Near East and the unidirectional flow of ideas out of this center to southern and Central Asia.

  If cattle were domesticated in Pakistan, why not sheep and goats along with them? The wild forms of all three animals were a part of the Mehrgarh environment. The efficiency of scientific reasoning, Occam’s razor, is best served by accepting a model indicating that local processes were at work.

  Barley and wheat are another matter. The case for the local domestication of barley is not so much an issue as that of wheat, since wild wheat has never been documented in the Afghan-Baluch region. But the region is a suitable habitat for wild wheat, and intensive botanical survey might change that observation. So, too, might paleobotanical /paleoenvironmental research demonstrate that wild wheat was present in the early Holocene and then became extinct. Both of these approaches represent serious research challenges, and both are testable.

  If the cattle, sheep, and goats were locally domesticated, something I take to be true for cattle and likely for sheep and goats, then it makes sense for us to think that the ancient peoples of the Afghan-Baluch region were also engaged in the domestication of plants. The sensible thing to do is to treat this statement and others like it as testable archaeological hypotheses and to not continue along, at the expense of all other ideas, with the old dogma advocating a culturally advanced Near Eastern “hearth” and the diffusion of domestication from it.

  There are indications that the development of food production and the domestication of all early plants and animals did not come to the Subcontinent by diffusion from the west. What is emerging is a notion that the old Near Eastern hearth of domestication was simply much larger than older hypotheses assumed. Rather than ending in the Zagros Mountains of Iraq/Iran, it spread all the way across the Iranian Plateau to the Indus Valley. The early Holocene peoples across this vast region, from the Mediterranean to the Indus, were engaged in the development of food-producing practices at that time. This area can be seen as a large interaction sphere in prehistoric times. This is the expanded nuclear zone for Near Eastern, Iranian, Central and South Asian domestication. The domestication of those pl
ants and animals on which Near Eastern, South, and Central Asian civilizations were founded seems to have taken place in this nuclear zone. Interaction within it may have been so intense and regular that future excavations will find no predominant early center of innovation within it, but that the ideas and products of early experiments with plants and animals were rapidly disseminated within the interaction sphere. The forces of cultural change and adaptation were regional, rather than local (Near Eastern, South Asian, Iranian, Afghan, etc.). Rich communication and sharing of ideas and products were essential ingredients in this process of cultural change. A useful, productive, successful innovation in one place may have been quickly adopted elsewhere, with no single region emerging as a clear and consistent “leader” in the process, at least not over the time scale used in prehistoric archaeology. (The means and mechanisms of these communications are outlined in other sections of this book.)

  There are two other points that should be made about Mehrgarh I. First, Mehrgarh is on the edge of the natural distribution of the potentially domesticable plants and animals, not within it. The inhabitants of this site had already gained sufficient control over their subsistence system that the plants and animals could be removed from their wild habitats and still survive. Second, without doubting the importance of Mehrgarh, and the very high quality of the French excavations there, we should recall that the earliest inhabitants of the site came there with a broad suite of fully domesticated food grains, including both wheat and barley. In this sense, Mehrgarh I is not a place where we have evidence of the earliest experiments with domestication. Mehrgarh might be compared to an early Pre-Pottery Neolithic B site in the Near East, with its mud-brick buildings and domesticated food grains. The really early experiments with food production on the Indo-Iranian borderlands are found closer to, even embedded within, the habitats of Braidwood’s constellation of potentially domesticable plants and animals. This is the mountains of Baluchistan, the Northwest Frontier, and Afghanistan, at such places as Aq Kupruk.